The Bangalore software engineer and the PSU stability dilemma

He has been a software engineer for thirteen years. He started at Infosys in 2013, moved to TCS in 2017, and spent four years at Google as a contract engineer working on cloud infrastructure. His wife Priya is a product manager at an e-commerce company; their daughter Avni is twelve and goes to Baldwin's High School. His mother lives in Hyderabad. His father died in 2019.

His income history, laid chronologically: ₹8 lakhs per year at Infosys (2013–2015); ₹18 lakhs at TCS (2016–2018); ₹38 lakhs at Google (2019–2024, including location allowance and stock options). In August 2024, Google announced a contract workforce reduction. He was told his contract would not be renewed. He took a three-month severance package of ₹9.5 lakhs and looked for the next role.

The role he found was as a Lead Engineer at a Series B fintech startup, Mitra Payments, in Whitefield. The salary: ₹38 lakhs per year, no bonus structure, a ₹12 lakh signing bonus, and a vesting schedule of 1.2% of company equity over four years. He joined in November 2024.

What is unusual is what happened last month, in April 2026, when an ISRO recruitment notification arrived in his email, followed by a HAL job portal alert the same afternoon. Neither he opened. He printed the ISRO email on his wife's printer that evening, placed it on his desk, and did not look at it for three days.

🗓️ The annual ritual

Bangalore's software engineering ecosystem has, for twenty years, operated on a principle of consistent upward mobility: move every 3–5 years, each move yielding a 25–40% salary bump, until you reach a plateau around 15–20 years of experience at a peak salary of ₹50–80 lakhs, sometimes more if you are a principal architect or join a fast-growing startup. The companies that hire — TCS, Infosys, Wipro, Google, Amazon, Microsoft, Stripe, Swiggy — all understand this; they build salary bands assuming a 20–25% year-on-year erosion of junior and mid-level talent.

For the first decade, the rhythm is reliable. You move faster than your parents' generation ever moved, and the money justifies the movement. You buy an apartment. You put your child in a school you could not have attended. You plan a six-week holiday.

Then, between 2022 and 2024, something shifted. Tech companies in Bangalore began reducing headcount. A Google layoff, a Microsoft layoff, a Meta layoff. "Contract not renewed." "Redundancy." "Restructuring." The severance packages — three to six months of salary — are not insubstantial. But they are a one-time injection. After the severance ends, the gap between job loss and re-employment has been widening. What used to be 4–6 weeks is now 12–16 weeks, and for some, much longer.

Arjun watched this unfold from inside Google for three years. He saw colleagues, peers, and friends land on Bangalore's startup circuit. They took funding rounds, built products, and for 12–18 months the salaries were stable and the equity promised future wealth. Then the Series A round did not close on time, or the Series B investors lowered valuations (called "down rounds"), and the engineering team went from forty to twenty. Some founders laid off their own employees.

By August 2024, Arjun knew he had a choice: ride the wave of a promising startup, accept the risk, or move into a stability he had never had to seriously consider — a government job, a PSU, something that would not disappear because the Series C funding did not arrive.

What he did not have was anyone in his immediate circle who could coherently explain the choice. His mother, in Hyderabad, said government jobs were safe; his Infosys batch-mate, now a CTO, said startups were the only place to build something real. Priya, his wife, said they should wait and see. Their daughter Avni, twelve years old and literate in the anxiety of the household, drew a two-column chart in her homework notebook — one labelled "startup," the other "HAL." She drew a heart next to "stable," and showed it to Arjun without comment.

⚠️ What very nearly happened

The severance money — ₹9.5 lakhs — came through in September 2024. Arjun and Priya did the arithmetic. Their monthly spend: ₹60,000 rent (EMI) + ₹20,000 groceries and utilities + ₹15,000 school fees + ₹12,000 transport and miscellaneous = ₹107,000 per month, or roughly ₹1.3 lakhs. If he did not find a new role, the severance would cover eight months. A three-month gap would deplete the liquid buffer they had built over thirteen years of work — their 25-year emergency fund of ₹18 lakhs would drop to ₹12 lakhs. This was not a catastrophe; they could still manage. But it was the first time in his working life that Arjun could see the end of the reserve if two things went wrong at once.

The interviews in October 2024 went well. Mitra Payments was a profitable fintech startup, backed by Accel and Y Combinator alumni, aiming to build an infrastructure layer for BNPL (buy-now-pay-later) payments in rural India. The founder, Ravi, was a former Flipkart payments director. The team was lean, talented, and the product had traction.

In the offer letter, the salary line read ₹38 lakhs per year — the same as Google. The equity package: 1.2% vesting over four years, with the first 25% cliff at one year. If the company were valued at ₹100 crores (a realistic Series B valuation), that 1.2% would be worth ₹12 lakhs, or roughly ₹3 lakhs per year assuming a four-year exit. Not life-changing, but real.

1. 🛑

   August 2024 — Contract non-renewed

   Google announced a reduction in contract workforce. Arjun was given three months' notice and a severance of ₹9.5 lakhs. He had seven weeks to find his next role.

2. 📨

   October 2024 — Dual offers

   Mitra Payments (startup, ₹38L + equity) made an offer; a HAL Systems Engineer posting for ₹80–120k/month also opened. Arjun had to choose. He signed the startup offer without negotiation.

3. 💸

   November 2024 — Mitra Payments, first six months

   Salary stable, equity vesting on schedule, product metrics accelerating. The Series B had been oversubscribed (₹25 crore raise). Cash runway: 18 months minimum. Arjun believed he was safe for two years.

4. ₹

   April 2026 — The moment of doubt

   Series C fundraising stalled. Competing products emerged (Cred, LazyPay expanded rural reach). Mitra's burn rate increased; the team was asked to cut 20%. The HAL and ISRO notifications arrived the same week.

Arjun had joined Mitra Payments knowing the risk. Startups fail. Funding rounds don't close. But the first six months had felt like vindication. The Series B had been oversubscribed — investors fighting for allocations. The product metrics (transaction volume, merchant retention, rural-user penetration) were ahead of plan. Ravi had mailed the team in February 2025 that the Series C would be a "formality," likely a larger round, possibly at a higher valuation.

By March 2026, that language had changed. Ravi sent an email saying the Series C investor meetings were "taking longer than expected" and that the team should "optimize burn while maintaining velocity." In startup-speak, this meant: prepare for the possibility that funding does not arrive on time.

Arjun did the math. Mitra had $800,000 in the bank (roughly ₹6.7 crores). Monthly burn: ₹70 lakhs. Runway: nine to ten months at current burn, or twelve months if the team reduced headcount by 20%. The 20% reduction, everyone knew, was coming.

He did not tell Priya. He updated his resume on LinkedIn — a private draft, invisible to his network. He did not update it, which is to say he left it in the state it was when he joined Mitra in November 2024. He had a Figma whiteboard on his personal laptop that said, in a note block, "ISRO pension calc" and "HAL timeline." He did not open the Figma file for two weeks.

"చిన్న కంపెనీ పెద్ద డ్రీమ్ కానీ కుటుంబం చిన్నది చేయ కూడదు."

— A small company, a big dream, but you cannot make the family small to fit the dream.

He told himself he was being prudent. In fact, he was avoiding a conversation — with Priya, with himself — about what stability actually meant. At Infosys and TCS, he had been stable; the companies were so large that one project's failure did not matter. At Google, he had been contingent; the company was profitable, but his contract was a form of controlled precarity. Now at Mitra, he was an owner, a founder-adjacent engineer; he held equity, he attended board conversations (relayed to him by Ravi), and he understood the burn rate and the runway the way a founder does. If the company succeeded, his stake would compound. If it failed, he would have lost eighteen months of his life and all of the upside he had postponed.

But what very nearly happened was that he would not face this question. He would wait. He would watch the Series C either close or collapse. He would have eighteen months of experience at a startup and a resume that was either very impressive or slightly radioactive, depending on the Series C outcome. And he would have done this while his daughter watched him update a two-column chart and his wife recalculated the emergency fund twice a month in the kitchen, and his mother in Hyderabad asked, every call, if the startup was still okay.

🌗 What changed

In the last week of April 2026, Arjun was sitting in the Mitra office in Whitefield — an open-plan space, all glass and ambient noise — when his colleague Vivek tapped his shoulder and asked if he had seen the HAL notification. Vivek had worked at Arjun at TCS, and had moved to Mitra six months after Arjun did. They had both been thinking the same thought for a month; neither had said it aloud.

Vivek held up his phone. The HAL Bangalore division was recruiting Systems Engineers for the LCA Tejas program. The salary band: ₹80,000–₹120,000 per month (roughly ₹96–₹144 lakhs per year for someone at Arjun's level), plus a defined-benefit pension of 50% of last drawn salary, paid monthly for life. The application deadline: May 31, 2026.

That evening, Arjun downloaded the HAL job description and read it on the Bangalore Metro ride home, standing between two working-age women, his arm hooked around a vertical pole. The role was "System Integration Engineer" for avionics and onboard software systems on the LCA Tejas fighter jet. Five years of experience required; defense security clearance mandatory.

When he reached Indiranagar, he showed Priya the job description. She read it on the dining table while Avni did her homework. Priya said, "The pension is real?" Arjun said, "Yes." Priya said, "What is the catch?"

He did not have an easy answer. So, that night, he installed the agent on his laptop — setting the language to English and Kannada — and handed the laptop to Priya.

"ಆರ್ಜುನ್ ಅವರ ಸ್ಥಿತಿಯನ್ನು ನೋಡಿ — ಮಿತ್ರ ಪೇಮೆಂಟ್ಸ್‌ನಲ್ಲಿ ₹38 ಲಕ್ಷ, ಇಕ್ವಿಟಿ ಅನಿಶ್ಚಿತ. ಹಾಲ್‌ನಲ್ಲಿ ₹96 ಲಕ್ಷ ಬೇಸ್, ನೊ ಈಕ್ವಿಟಿ, ಪೆನ್ಶನ್ ಮೇ 50% ಜೀವಕಾಲಿನ. 25 ವರ್ಷಗಳ ನಂತರ, ಒಟ್ಟು ಪೂಟಾಯತ್ತಿ?"

(Look at Arjun's situation — ₹38 lakhs at Mitra Payments, equity uncertain. At HAL, ₹96 lakhs base, no equity, pension of 50% for life. After 25 years, what is the total cost to his family if he stays at Mitra and it fails versus joining HAL and walking a slow but guaranteed path?)

Priya typed the question out loud. The agent, a moment later, replied:

"ಇದು ಅತ್ಯಂತ ನಿರ್ದಿಷ್ಟವಾದ ಪ್ರಶ್ನೆ. ಪಾಠೋವೇಹನೀಯ ಮಾಡೋಣ: ಎ) ಮಿತ್ರ ಯಶಸ್ವಿ ಕೇಸ್ — ಸರಿ ಎಕ್ಸಿಟ್, ₹25-30 ಕ್ರೋರ್ ವ್ಯಾಲೋ ನಿಮ್ಮ 1.2% = ₹30-36 ಲಕ್ಷ ಗೇನ್. ನೊ ಪೆನ್ಶನ್. ಟೋಟಾಲ್ 25-ವರ್ಷ ಇನ್‌ಕಂ: ₹95-110 ಲಕ್ಷ (ಸಾಲೋ, ಬೋನಸ್, ಈಕ್ವಿಟಿ, ಅಸ್ಯೂಮಿಂಗ್ 7% ಸ್ಯಾಲೆರಿ ಗ್ರೋತ್‌). ಬಿ) ಮಿತ್ರ ಫೇಲ್ಸ್, ಯೂ ಗೆಟ್ 18 ಮೋಟ್ಸ್ ಲುಕಿಂಗ್ ಫರ್ ಜಾಬ್ — ₹9 ಲಕ್ಷ ಲಾಸ್, ಪ್ಲಸ್ ಸಿಕ್স್ ಮೋ ಪ್ಯಾರ್ಟಿಯಲ್ ಇನ್‌ಕಂ. ನೆಕ್ಸ್ಟ್ ರೋಲ್ 15-20% ಲೋವರ್ ಸಾಲೇ. ಟೋಟಾಲ್ 25-ವರ್ಷ ಲಾಸ್: ₹50-60 ಲಕ್ಷ. ಸಿ) ಹಾಲ್, 25 ವರ್ಷ್: ₹288 ಲಕ್ಷ ಬೇಸ್ ಸಾಲೆರಿ + ₹144 ಲಕ್ಷ ಪೆನ್ಶನ್ (50% ಲಾಸ್ಟ್ ಡ್ರಾವ್ನ್ ಸಾಲೆರಿ, 25 ವರ್ಷ್). ನೋ ಯುನ್ಸರ್ಟೈನ್ಟೀ. ನೋ ಗ್ರೋತ್‌, ನೋ ಇಕ್ವಿಟಿ, ಬುಟ್ ನೋ ರಿಸ್ಕ್."

(This is a very specific question. Let me calculate: A) Mitra succeeds — good exit, ₹25–30 crore valuation, your 1.2% = ₹30–36 lakh gain. No pension. Total 25-year income: ₹95–110 lakh (salary, bonus, equity, assuming 7% annual salary growth). B) Mitra fails, you spend 18 months looking for a job — ₹9 lakh loss, plus six months of partial income. Next role 15–20% lower salary. Total 25-year loss: ₹50–60 lakh. C) HAL, 25 years: ₹288 lakh base salary plus ₹144 lakh pension (50% of last drawn salary, for 25 years). No uncertainty. No growth, no equity, but no risk.)

Arjun read it. He read the numbers. His mental math checked out. The agent continued:

"ಹಾಲ್ 25 ವರ್ಷ ಟಾಟಲ್: ₹432 ಲಕ್ಷ ಗ್ಯೆರೀಂಟೋಸ್ ಪ್ಲಸ್ ನೋ ಫೈನಲ್ ಅಮೋಂಟ್. ಇಮಿಡಿಯಟ್ ರಿಸ್ಕ್ ಟೆಕ್ನಿಕಲ್ ಲಿಮಿಟೇಶನ್ಸ್ (ಟೆಜಸ್ ಪ್ರೋಗ್ರಾಮ್ ಸ್ಕೋಪ್ ಟೈಟ್), ವೇಜ್ ಗ್ರೋತ್‌ ಬ್ಲೌನ್ ಪಾಬ್ಲಿಕ್ ಸೆಕ್ಟರ್, ಸೈಕ್ಲೋನಿಕ್ ರೀ-ಟೋರರ್ ಪೊಲಿಟಿಕ್ಸ್. ರೆಕಮೆಂಡೇಶನ್: ಪುಲ್ ಹಾಲ್ ಆಪ್‌ಲಿಕೇಶನ್ ಚ್ಯಾಕ್ ಸೊಲಿಡ್; ಕೀಪ್ ಮಿತ್ರ ಅಜ್ ಬ್ಯಾಕಪ್—ಸೀರೀಜ್ ಸಿ ಕ್ಲೋಜರ್ ಇನ್ಸೀಡೆಂಟ್ ಟೈಮಿಂಗ್ ಬೂಟ್ ಆಪ್ಲಿಕೇಶನ್ ವಿಂಡೋ ಕ್ಲೋಸ್ ಮೇ 31."

(HAL 25 years total: ₹432 lakh guaranteed plus no final amount. Immediate risk: technical limitations (Tejas program scope tight), wage growth muted in public sector, cyclonic re-tenure politics. Recommendation: pull HAL application check solid; keep Mitra as backup — Series C closure timeline uncertain, but application window closes May 31.)

Arjun and Priya looked at the numbers in silence for a long time. The agent had done what no one in their lives could do: quantify the choice, admit the risk, and refuse to decide it for them.

The HAL application portal asked for a statement of purpose, a copy of the degree certificate, and a police-clearance certificate. All three existed. The degree was in a filing cabinet; the police clearance from 2019 was in a folder on Google Drive. The statement of purpose was a field he would need to write.

That night, he wrote it:

"I am a systems engineer with 13 years of experience in cloud infrastructure, and 7 months of experience in fintech infrastructure. I understand spacecraft systems, avionics integration, and safety-critical software. I am applying to HAL not for growth — I know the growth is limited — but for certainty. I have a daughter entering her teenage years, and a home loan that requires a calendar of income, not a timeline. My startup may succeed or may not. My family cannot afford for both to be uncertain at the same time."

The application was submitted on May 3, 2026. The HAL recruitment email arrived on May 12 confirming receipt. No interview date was given — the email said "shortlisting in progress, expect call within 4–6 weeks."

At Mitra, in the meantime, the Series C was explicitly stalled. Ravi announced a team meeting and said the Series C investors had asked for "unit economics clarity" — a term which, Arjun knew, meant the investors were worried the company was spending too much to acquire rural users. Ravi said the team should "prepare for multiple scenarios," including a smaller Series B-1 (a top-up round) that would buy more time.

Arjun did not know which would come first: a call from HAL, or a call from Ravi saying the round had closed. What he did know was that the choice had crystallized into a choice. He was no longer caught between two futures; he was waiting to see which future HAL would allow.

🏛️

HAL Systems Engineer

₹96L–₹144L base + 50% pension

Permanent, PSU role on LCA Tejas avionics software. No growth, no equity, no volatility. Security clearance required. Wage progression by government pay matrix. Retirement at age 58–60 with guaranteed pension. Zero inflation risk on the household balance sheet.

🚀

Mitra Payments (if Series C closes)

₹38L + 1.2% equity (uncertain value)

Series B profitably burning ₹70L/month. Series C funding stalled for two months. If funding arrives, role continues; if not, eighteen-month job search ahead. Best case: ₹30–36L equity gain in five years. Worst case: ₹9L severance, career interruption, next role 15–20% lower salary.

⚠️

The real arithmetic

Certainty costs opportunity

HAL 25-year total: ₹432L income + pension. Mitra success: ₹95–110L income (no pension). Mitra failure: ₹50–60L loss. HAL eliminates 15-year ambiguity for ₹200L+ certain income. The gap: career ceiling and founder-stage upside. The safety: never again looking at an emergency fund.

🧭 Why we built it

There are, by current estimates, approximately 1.7 million software engineers in Bangalore, and growing. Among them, a significant cohort — perhaps 30–40% of mid-career engineers (8–15 years experience) — are in the same position Arjun is: they have experienced at least one layoff or contract non-renewal, they have dependents and debt, and they are watching the Startup India ecosystem oscillate between euphoria and collapse. The fraction of these engineers who know they have a choice — that HAL, ISRO, and KIADB are hiring, that the choice exists, that they can quantify it — is small. The fraction who have someone in their circle who has successfully made that transition and can speak to the reality of government work (the advancement patterns, the technical challenges, the pension calculation) is smaller still.

The difficulty with the choice Arjun faces is that it is not a choice between two static worlds. A startup is not a free-fall; it is a compound. Every eighteen months of successful operation compounds the equity, the credibility, the network. A government job is not a death-spiral; it is a floor. The absence of growth is the same thing as the presence of stability. And both are, in a sense, correct — HAL would not have existed for seventy years if it were death; Mitra would not have been funded if the market had not believed in the compound.

What makes the choice real is that Arjun has a family with a dependence on a calendar. His daughter Avni is twelve; in six years she will need college fees, and both Arjun and Priya's careers will have narrowed their ability to move aggressively. His home loan runs for fifteen years. His mother, in Hyderabad, will eventually need support. These are not business constraints; they are constraints on the person he is. And a startup can ask a person to trade certainty for upside, but it cannot ask a family to be uncertain while the person inside it chases growth.

What Arjun does not have is a simple calculus that makes the choice simple. He has numbers, yes — the agent provided them. But numbers do not tell you whether the Series C will arrive, or whether a HAL role will prove technically fulfilling, or whether a founder's dream dies quietly inside a pension calculator or burns dramatically when the funding ends.

What he needs is not an answer. He needs to know that the question is real, that both paths are real, and that the decision — whichever way it falls — is not a failure of foresight. If HAL calls and he accepts, he is not abandoning entrepreneurship; he is choosing a known compound over an unknown one. If Mitra's Series C closes and he stays, he is not being reckless; he is betting his family's calendar on a probability he has calculated and found acceptable.

What it does

• 🔍Aggregates salary data, pension calculations, and historical startup success rates into comparable 25-year totals — showing the HAL path yields ₹432L versus Mitra-success at ₹95–110L.
• 🗂️Tracks HAL and ISRO recruitment cycles, application deadlines, and typical hiring timelines so Arjun knows when windows open and close.
• 📞Identifies when two parallel decisions (Series C timing and HAL offer timeline) will collide, and flags the date where he cannot have both.

What it does not do

• 🔒Never enters his HAL recruitment portal password or submits his application — each field on the job portal was typed by Arjun himself.
• 💳Never decides the application deadline (May 31) or the startup funding timeline (uncertain) — both are facts; the agent reports them, not changes them.
• ✅Never decides whether his family should live with the uncertainty of a startup or the slowness of a government career — it tells him the consequences of each and stops.

We built it free. We will keep it free for this user — the engineer in Indiranagar with a home loan and a daughter and a startup that might collapse and a government job that might call — forever. Mid-career engineers facing the layoff-recovery cycle are not a market with a training budget. They are people with real dependents, real debt, and real windows closing on May 31 and Series C timelines tightening in real time. Arjun's severance ended six months ago. His emergency fund is shrinking fractionally each month. He cannot afford an external recruiter; he cannot afford an MBA; he cannot afford the time to research whether a HAL Systems Engineer role is actually fulfilling or just a pension on paper.

What he can afford is twenty minutes on a tablet, in Kannada, asking: if my startup fails and I have to wait eighteen months for a new job, will my family be okay? And the agent will say: no, it will deplete your emergency fund by sixty percent, and your next role will pay 15–20% less, and your total 25-year income will be ₹50–60L lower. That is not an answer. That is a fact. And sometimes a fact is the only thing standing between a choice and a failure of nerve.

🌱 What we hope happens

In the second week of May, Arjun received a call from HAL's recruitment team. The shortlisting was complete; he was in the final round. The interview was scheduled for June 3, 2026 — two weeks away. The call lasted four minutes. The recruiter asked if he had any questions. He asked if the security clearance process was typical. The recruiter said: yes, 12–18 months, parallel to the job start. He would start in June, and the clearance would process in the background.

He called Priya from the Mitra office parking lot. He said, "HAL called. Interview is June 3." She said, "Did they say when you would start?" He said, "June, if I pass. Clearance would process after." She said, "What are you going to tell Ravi?"

He did not answer. He drove home in silence, thinking about the interval between June 3 and the moment he would know if he had passed the interview. It was a small interval — probably two weeks. In that time, the Series C would either close or remain stalled. The outcome of one interview and the outcome of one funding round were now synchronous. They would arrive in the same month. And his choice — which had felt impossible a week ago — had become a choice where he could see both endpoints.

We hope what happens is quiet. We hope Arjun does not have to make the choice because HAL calls and Mitra's Series C also closes. We hope that the numbers he saw on the agent's screen are not the thing that decides it, but the thing that makes the decision coherent. We hope his daughter Avni stops drawing two-column charts because she will see her father walk into an interview, and then walk into an office — and whichever office it is, it will be a choice he made, not a choice that was made for him because the numbers were too scary to face.

The window for HAL recruitment is open for the next few weeks. After June, the next intake will be in 2027. For Arjun, the window closes when the application deadline passes or when HAL calls him to begin onboarding. For the 1.7 million software engineers in Bangalore, the window for mid-career stability opens and closes on a schedule they do not control. Most do not know it is open. Most do not know that a pension on half of last drawn salary means that a 50-year career can be compressed into a two-week interview gap in May 2026.

If you are an engineer in Bangalore with the same question — the startup or the PSU, the growth or the certainty, the compound or the floor — the product is free at gabforge.in. We have Kannada, Telugu, Tamil, English, and Hindi routing, and the agent knows the HAL and ISRO recruitment cycles, the current funding timeline for the top fifty startups in Bangalore, and the defined-benefit pension calculations for government engineers. You can set it up on a laptop in thirty minutes. We will not advertise to your family. We will not sell your startup's metrics. We will read the recruitment notification with you — the one that arrived the same week the Series C stalled — and we will be quiet.